HPC Scientist

Type of Requisition:Regular Clearance Level Must Currently Possess:Interim Secret Clearance Level Must Be Able to Obtain:Secret Suitability:No Suitability Required Public Trust/Other Required:None Job Family:Scientists Job Description:At GDIT, people are our differentiator. We are looking for an HPC (High Performance Computing) Computational Scientist to join our team at our Aberdeen, MD location. The team is working in a hybrid environment, primarily virtual but with occasional office requirements for work that cannot be completed virtually. We are a part of the Department of Defense (DoD) High Performance Computing Modernization Program (HPCMP) User Productivity Enhancement and Training (PET) initiative. PET enhances the capability and productivity of the HPCMP community through training, collaboration, tool and software development, and technology transfer. You will join a team of computational experts representing a critical mass of expertise in key areas of support for the HPCMP. These teams provide deep technical expertise in HPC that can be applied to a diverse range of scientific disciplines. Our teams provide deep technical expertise in HPC that can be applied to a diverse range of scientific disciplines. Sample areas include computational fluid dynamics (CFD); computational structural mechanics (CSM); computational electromagnetics and acoustics (CEA); forces modeling and simulation (FMS); integrated modeling and test environments (IMT); autonomy; high-performance data analytics (HPDA); and physics informed machine learning (ML)/Artificial Intelligence (AI). Duties may include the following;Software Refactoring to Optimize Performance for Emerging HPC ArchitecturesEmerging Hardware Exploration: Supporting the evaluation of emerging hardware for potential deployment in future HPC systems.Assist project sponsors with scientific research and or development studiesAnalyze problems and apply theoretical techniques to develop tools and software. Technical Skills should include a mix of the following:BS, 2+ years of experience; MS preferred.Expertise with multiple common scientific computing programming languages and programming models (e.g. C/C++, Python, Julia)Expertise with multiple common data analytics or machine learning tools (e.g. TensorFlow, Horovod, PyTorch)Expertise with legacy scientific computing programming languages (e.g., C, C++, Fortran)Experience with Linux on large scale systems.Extensive knowledge of HPC and parallel programming techniques (e.g. MPI, OpenMP), in multiple physical science domains.2+ years of experience with science or engineering programsMust be a US citizen and be eligible for a DoD Secret clearanceActive DoD Secret clearance preferred Scheduled Weekly Hours:40 Travel Required:Less than 10% Telecommuting Options:Onsite Work Location:USA MD Aberdeen Additional Work Locations: COVID-19 Vaccination: GDIT does not have a vaccination mandate applicable to all employees. To protect the health and safety of its employees and to comply with customer requirements, however, GDIT may require employees in certain positions to be fully vaccinated against COVID-19. Vaccination requirements will depend on the status of the federal contractor mandate and customer site requirements. We are GDIT. The people supporting some of the most complex government, defense, and intelligence projects across the country. We deliver. Bringing the expertise needed to understand and advance critical missions. We transform. Shifting the ways clients invest in, integrate, and innovate technology solutions. We ensure today is safe and tomorrow is smarter. We are there. On the ground, beside our clients, in the lab, and everywhere in between. Offering the technology transformations, strategy, and mission services needed to get the job done. GDIT is an Equal Opportunity/Affirmative Action employer. All qualified applicants will receive consideration for employment without regard to race, color, religion, sex, sexual orientation, gender identity, national origin, disability, or veteran status, or any other protected class.

Created: 2025-10-04